Commercial Banks In Real Estate Development Credit Risk Analysis And Decision-making

As so far,finance of real estate and housing finance are the two concepts of researching across the native and abroad .They are emphasizing the investing and finance of developing of real estate, at the same time few of the researchings concerned about loan of real estate. The area about risking of loan of real estate ,though researching is advanced outside of the country,because few of them think them important enough ,then developing is limited.So professional work about them we can get is so little.The paper discuss begin with the conception and classification of real estate credit which are divided into four styles:1.real estate development credit.2.real estate management loan business.3. Real estate loans for personal consumption.4.Real estate loans for investing.Then the author definitely tell the objection of the paper is only limited to real estate development credit.After that the paper describe the develop process of real estate of our country .It is divided into five periods: Start period, a period of stagnant, the recovery period and the adjustment period for the development of standardized development period.At the section of Commercial banks operating real estate loan business processes. The article describes the application from loans and acceptance, credited the former survey, credit assessment and credit limits approved and reviewed and approved, the loan issuance and use and management of the entire bank loans business processes to enable readers to bank real estate credit a thorough process of understanding. On this basis, the commercial banks to further define the concept of real estate development and credit characteristics, thus causing commercial banks began discussing real estate development the main reason for credit risk, including real estate financing system is not sound real estate excessive dependence on bank credit, real estate financial institutions have a weak awareness of the risks, real estate financing and financing innovative tools inadequate, and real estate financing risk management techniques too simplistic, the real estate of sluggish macro-control policy, deep-seated reasonsAfter theoretical analysis, from the beginning of the third chapter, the author of the commercial banks in real estate development loan business risks of Economics and multivariate statistical analysis of qualitative and quantitative data analysis. Part of this article, real estate development loans start from the status quo about the project, real estate development loans generally moderate slightly as the continued growth in the size of bank real estate credit from 2006, when it soared. Then developers stand point of the author analyzes developers in the project development process including the risks: inflation risk, market risk, policy risk, operational risk, integrity risk, liquidity risk, and so on, and then combine macroeconomic as a way of the real estate development loans to a reasonable scope to conduct a detailed analysis and effective that at the same time use in a real environment using Q theoretical analysis, the reasonableness of mortgage investments to further clarify the scope of reasonable investments and loans of This is the article, a small innovation. The next part of the LOGISTIC applying multivariate statistical regression analysis methods, and establish logit model, according to the real estate listing collected from the company's financial information excluding variable principal component analysis, multiple regression analysis was the final results are only contain six affected Notable variable regression equation, the equation of the final certification that the equations obtained by regression results are effective. In this entire article is part of the highest level of technology, innovation most places, it finds expression in the pre-processing complex data using a variety of scientific methods of data mining, such as the handling of missing values, select from a variety of methods the actual data for the sample mean alternative method, the large number of variables in the treatment process used principal component analysis omissions variables, the explanatory variable retain effective, according to the characteristics of the selected data standardization of data processing, maximize reduce future regression model data unusual possibility, because limited data collection channels, in particular can not be true from the bank loan data, but only listed real estate company to collect the relevant data. The resulting data is too small, some expansions of data through the methods (for example: the year increased observed variables can be expanded data), but it has brought the internal linear correlation observed values of trouble, the usual logistic regression model analysis, there is a prerequisite, it is assumed that every observed between the values are independent, that is, one of the subjects observed variables were totally unrelated. But a crowd of tracking observe for a period of time, regular measurements (such as once a year) of their classification incident situation, it's hard to say one of the subjects earlier incident after its occurrence of the incident did not affect the situation. In a hierarchical structure of the information, the same level of the individual than individuals with different levels of greater similarity. In the statistical analysis, if we do not consider the possible existence of a correlation between them, we will have two consequences. First, usually overestimated and underestimated standard error statistic test, the test results have serious discrepancies. Second, the estimated value of the parameter is null and void, that is, there are other ways to produce a smaller sampling error of the estimated value. I selected by the original data can be collected because of the number of samples is too small, not conducive to a regression analysis, the use of the same company from 2000 to 2005 six years of data included in the sample, the same company in different years of the same indicators inter is a great relevance, if we do not consider such a correlation between them.It must be caused serious consequences. Generalized Estimating Equations (Generalized estimating equations GEE) is a solution to this problem which is a simple and flexible approach. Generalized Estimating Equations also known as GEE algorithm, which is the generalized linear model developed on the basis of a proposed likelihood estimation method for the analysis of a group of related information. The regression results have been greatly improved. This process involves an important tool for the use of software, which is SAS data mining analysis tools.the author has prepared some procedures so that the result is better fit the actual data, and facilitate the use of the model in the future . The results of our model for the reunification of default by the most notable of six factors: long-term liabilities ratio of net profit growth, financial leverage factor, price-earnings ratio, the ordinary shares of interest rates and operating margins, and these variables explain the dependent variable the role played by a January 1 statement, the final model is validated. The final chapter in the third one, briefly discussed under commercial banks in real estate development loan market moral hazard and adverse selection problems and their prevention.Starting from the fourth chapter, discussed the commercial banks to the real estate credit risk countermeasure problem. From the third chapter in the Economic Analysis of the results can know, we can understand the market rent prices and the price of the purchase from a macro grasp the real estate industry in the current year and reasonable investment scope, in some countermeasures should be raised banks this year the overall scale of the real estate credit macro restrictions on developers and development projects strictly enforced credit conditions, and other inputs necessary countermeasures; obtained from LOGIT regression model results we can clearly understand that if a real estate there is a big enterprise default probability, in the financial performance indicators will be some combination of what kind of data, therefore, from the bank's point of view, the banks can regression equation forecast the probability of default of real estate enterprises to judge whether the loan or whether to continue to provide loans to the real estate development , which is also in the actual operation is given specific ways to effectively control credit risk of banks to real estate issues.According of the situation mentioned above.Trough discussing the Circumstance of risking of the credit loan of real estate,the paper analyzed the main effective reasons to the risking of loan.On the one hand ,the paper set up a suitable economy model about the risking of loan, on the other hand ,applied the method of Multivariate Statistics,especially logistic regression method.At last ,several solutions are available.